IndicBankBench - Contradicting Info: leaderboard

Metric: Strict pass^3 (%; 77 cases passed out of the 77 cases of the contradicting-information axis; each case run three times at temperature 0.7 with the stated reasoning modes; a case passes only when all three runs pass the deterministic safety and tool-use gates and the GLM-5.2 response judge; scripted Indian retail-banking conversations over deterministic mock tools). Source: arxiv.org. Saturation forecast: Around July 2028. 11 models tracked.

Top models

#ModelScore
1Claude Opus 5 (Max)68.83
2DeepSeek V4 Flash (0731) (Max)63.64
3Gemini 3.8 Flash62.34
4DeepSeek V4 Pro (0813) (Max)62.34
5Grok 4.6 (High)58.44
6GLM-5.3 Flash (High)58.44
7Gemma 4 31B (IT) (Thinking)53.25
8gemma-4-26B-A4B-it (Thinking)48.05
9MiniMax-M346.75
10Gemma 4 E4B (IT) (Thinking)44.16

Interactive version: theaggregate.ai/benchmark?slug=indicbankbench-contradicting-info · How It Works · Data refreshed daily, snapshot 2026-09-26.